An annotated database aimed at performance evaluation of algorithms for ST-T change analysis

Taddei, Alessandro; Biagini, Andrea; Distante, G.; Marchesi, Carlo; Mazzei, M. G.; Pisani, P; Roggero, N.; Zeelenberg, C.

doi:10.1109/cic.1989.130497

Cited by 8 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The database represents the results of an international effort for defining a standard in the analysis of ST-T changes in the ECG's waves from ambulatory recordings (AECG). The project was founded by the European Community and by the European Society of Cardiology [14], [15]. Thirteen research groups from eight European countries (coordinated by the Institute of Clinical Physiology (National Research Council) in Pisa (Italy) and by the Thorax Center of Erasmus University in Rotterdam (Netherlands)) provided the AECG tapes and annotated beat by beat the selected 2-channel recordings, each of 2 hours in durations.…”

Section: Materials and Proceduresmentioning

confidence: 99%

See 1 more Smart Citation

An Adaptive Learning Algorithm for ECG Noise and Baseline Drift Removal

Esposito

D’Andria

2003

Neural Nets

View full text Add to dashboard Cite

Abstract. Electrical noise and power line interference may alter ECG morphology. Noise reduction in ECG is accomplished applying filtering techniques. However, such filtering may mutate the original wave making difficult the interpretation of pathologies. To overcome this problem an adaptive neural method able to filter ECGs without causing the loss of important information is proposed. The method has been tested on a set of 110 ECGs segments from the European ST-T database and compared with a recent morphological filtering technique. Results showed that morphological filters cause inversions and alterations of the original signal in 65 over 110 ECGs, while the neural method does not. In 96% of the cases the signal processed by the network is coherent with the original one within a coherence value of 0.92, whereas this values for the morphological filter is 0.70. Moreover, the adaptability of the neural method does not require estimating appropriate filter parameters for each ECG segments.

show abstract

Section: Materials and Proceduresmentioning

confidence: 99%

“…Neural network for signal processing have suggested by several authors [11], [12], [13]. However, in ECG processing, their use was limited to classification tasks (see [14] and [15] among others). The present work proposes a filtering techniques based on an adaptive learning algorithm.…”

Section: Introductionmentioning

confidence: 99%

An Adaptive Learning Algorithm for ECG Noise and Baseline Drift Removal

Esposito

D’Andria

2003

Neural Nets

View full text Add to dashboard Cite

show abstract

“…A further study [8] documented that the reliability of the computer presentation of the time series of some ST-T changes parameters, was higher than in conventional analyzers. The results of both studies have been implicitly confirmed during the development of the European ST-T Database [2]. The cardiologists participating to the project, found that the measurements of the ST-T, reported on a computer print-out together with the original ECG signal, were very reliable.…”

Section: Principal Component Analysis Of the Ecgmentioning

confidence: 66%

“…The experience gained This work has been partially supported by a grant of REMCO-CARDIOLINE. in that study showed us that, in some circumstances, the delay between electrical and mechanical events could make difficult the interpretation of the analysis and that a reference database of annotated signals would be necessary to assess the performance of the method. Recently such a database has been developed [2] reporting the annotations of ST-T changes, as manifested on two leads ECG, while an annotated database for pressure signal is still lacking.…”

Section: Introductionmentioning

confidence: 99%

Monitoring patient status through principal components analysis

Buemi

Macerata²,

Taddei³

et al.

[1991] Proceedings Computers in Cardiology

Self Cite

View full text Add to dashboard Cite

Signal monitoring produces a large amount of redundant data, and a compact data presentation technique would be highly desirable. Principal component (PC) analysis allows the definition of a transform with optimum coefficients. The PC's of several time series of the features, extracted from the ECG or the arterial pressure, are calculated.With the assumption that the statistical structure of intra patient data is going to be stable, the basis functions of the transform are captured once, during a basal interval. A function has been derived from the two jirst PC's, which represents an evidence function to be used both for a compact visual presentation and for the design of an algorithm for automatic episode detection. The evaluation of the visual presentation has been based on the annotated signal of the European ST-T Database.

show abstract

“…The initiation of a standard approach to the detection and interpretation of ST-T changes was made by the "Concerted Action" on Ambulatory Monitoring, set up by the European Community in 1985 [5]. As reported at previous Computers in Cardiology conferences [6,7], the ECG project concentrated on setting up an annotated database of ST-T changes. The Concerted Action ended in 1986 after setting up a small prototype database.…”

Section: Introductionmentioning

confidence: 99%